APPARATUS FOR CLASSIFICATION OF MOTION IMAGINARY EEG SIGNALS AND METHOD THEREOF

摘要

The present invention relates to a motion imagined EEG signal classification device and method thereof. According to an embodiment of the present invention, there is provided a method of imagining an EEG signal using an apparatus for imagining an EEG signal, comprising: receiving an EEG signal obtained for a predetermined time through an operation imagination experiment; Dividing a frequency of the EEG signal to correspond to the frequency range by using a filter bank configured by equally dividing a frequency range into a predetermined number; Extracting and normalizing feature information for each frequency range of the EEG signal through a normalized common section pattern (R-CSP) filter; Selecting a predetermined number of feature information pairs for constructing a parameter set for an ensemble from the extracted feature information using a MIBIF (Mutual Information Based Individual Feature) algorithm; And classifying the imaginary motion of the EEG signal using an ensemble technique based on the selected feature information pair. As described above, according to the present invention, a normalized common spatial pattern (R-CSP) filter is applied to the EEG signal passing through the filter bank and ensemble to accurately classify the characteristics of the motion as expected, thereby improving the performance. have.